Localization Algorithm Research For Internet Of Things

Internet of Things(IoT) is an intelligent network which connects all kinds of things through the information sensing devices in accordance with agreed protocols for the purpose of exchanging information and communicating. It achieves the goal of locating, tracking, intelligent identifying, and managing things. IoT has broad application prospects, such as intelligent robot, transportation, intelligent buildings, and other fields. Precise positioning of nodes is the key problem of location service in IoT. In many situations the relevant data with the location service application is worthwhile when accurate location information of nodes is acquired. So accurate positioning in IoT plays an important supporting role in realizing the location-based services and improving life quality, which has enormous social value and commercial value.On the basis of research achievements in localization algorithm of IoT, we improved the localization algorithm to further enhance the positioning precision. The conventional Taylor series expansion is applied to solve distance equations between unknown nodes and anchor nodes. Location information is not comprehensive enough to result in lower positioning accuracy. Thus, firstly a new positioning model based on multivariable Taylor series expansion was established, which considers distances between unknown nodes. Then, two localization algorithms based on the new positioning model were proposed. One is based on trilateration localization algorithm. It firstly uses trilateration method to get initial position of unknown nodes. Then, the estimated positions of unknown nodes are calculated by the weighted least squares method. The other is based on MDS algorithm. After initial values of unknown nodes position are obtained by MDS algorithm, least square method is adopted to find the solution of the positioning model. To evaluate the performance of these algorithms, Cramer-Rao lower bound(CRLB) of positioning result was derived. Simulations test the impact of different distance measurement error and anchor node number on positioning error, and its cumulative distribution function. Simulation results show that these proposed algorithms can effectively improve positioning accuracy and their positioning error are very close to the CRLB. Finally, a simulation platform for node localization of IoT was designed and implemented using MATLAB. The user can set the parameters such as the number of anchor nodes, the number of unknown nodes through a visual interface provided by the platform. Meanwhile, the platform can analyze positioning performance intuitively.